Sonic Software, the inventor and leading provider of the enterprise service bus (ESB), has released Sonic ESB 5.5. Sonic ESB makes it possible for companies to build an event-driven, service-oriented architecture (SOA) that can adapt to ever-changing business requirements.
Sonic ESB 5.5 incorporates Sonic Continuous Availability Architecture (CAA) to deliver highly available communications between applications in an enterprise SOA.
Sonic Continuous Availability Architecture dramatically reduces the time required for the communications infrastructure of the ESB to resume operations seamlessly after hardware, software or network failures, and guarantees that transactions are not lost or rolled back. Key capabilities include hot failover to ensure the integrity of in-process transactions; and an 'out-of-the-box' software-based configuration, eliminating the need to configure and deploy specialised hardware.
"Continuous availability is the latest in a series of critical capabilities that enterprise-grade ESBs must provide," said Gordon Van Huizen, CTO of Sonic Software. "As the market embraces ESBs as the future of integration, this release extends our advantage, and further refines our definition of best-of-breed ESBs."
"A plain ESB is sufficient for implementing a new, basic SOA or event-driven application, but it does not incorporate all of the features that are helpful in more-sophisticated applications or in application integration scenarios," said Roy Schulte, vice president and research fellow at Gartner, Inc. "Architects must examine ESB and integration suite architecture at a detailed level to be able to detect the subtle but important differences in product strengths and limitations."
Since Sonic Software delivered the first ESB to market in March of 2002, the technology and term have gained widespread adoption across enterprise IT shops and the software industry alike. Today, the basic building blocks of an ESB are fairly well established: messaging; Web services; intelligent routing; and data transformation. Though necessary, these basic capabilities are not sufficient to fully support enterprise service-oriented architecture deployments in practice.
Best-of-breed ESBs include:
Distributed Services Architecture - provides a flexible foundation that supports an event-driven enterprise SOA. Required functionality: distributed, dynamic deployment, management and monitoring of remote services from a centralised administrative console; and a configuration-driven implementation that allows each service to be independently scaled, reconfigured and redeployed without disrupting other applications or operations.
Enterprise-class Communication Backbone - for robust, scalable and secure communications across the extended enterprise. Required functionality: guaranteed message delivery; global scalability; high availability; a flexible security infrastructure; and support for Web services.
Intermediary Services and Adapters - provide a broad set of distributed services to facilitate application data exchange. Required functionality: distributed intelligent routing; XML transformation; support of centralised auditing and logging; and adapters for service-enabling third party and legacy applications and interfacing to specialised industry protocols.
"As other vendors bring first-generation ESBs to market, there is a risk that these offerings will lack a coherent architecture and usability in real world deployments," continued Van Huizen. "We challenge vendors to move beyond the basic architectural concepts and provide the full range of capabilities necessary to ensure flexibility, reliability and scalability, including a consistent architecture, deployment framework and toolset."